Commit Graph

157352 Commits

Author SHA1 Message Date
Alexander Graf
ff092385e8 KVM: SVM: Implement INVLPGA
SVM adds another way to do INVLPG by ASID which Hyper-V makes use of,
so let's implement it!

For now we just do the same thing invlpg does, as asid switching
means we flush the mmu anyways. That might change one day though.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:58 +03:00
Alexander Graf
3c5d0a44b0 KVM: Implement MSRs used by Hyper-V
Hyper-V uses some MSRs, some of which are actually reserved for BIOS usage.

But let's be nice today and have it its way, because otherwise it fails
terribly.

[jaswinder: fix build for linux-next changes]

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:58 +03:00
Alexander Graf
0367b4330e x86: Add definition for IGNNE MSR
Hyper-V accesses MSR_IGNNE while running under KVM.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:58 +03:00
Avi Kivity
b3dbf89e67 KVM: SVM: Don't save/restore host cr2
The host never reads cr2 in process context, so are free to clobber it.  The
vmx code does this, so we can safely remove the save/restore code.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:58 +03:00
Avi Kivity
d3edefc003 KVM: VMX: Only reload guest cr2 if different from host cr2
cr2 changes only rarely, and writing it is expensive.  Avoid the costly cr2
writes by checking if it does not already hold the desired value.

Shaves 70 cycles off the vmexit latency.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:57 +03:00
Jan Kiszka
681405bfc7 KVM: Drop useless atomic test from timer function
The current code tries to optimize the setting of
KVM_REQ_PENDING_TIMER but used atomic_inc_and_test - which always
returns true unless pending had the invalid value of -1 on entry. This
patch drops the test part preserving the original semantic but
expressing it less confusingly.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:57 +03:00
Jan Kiszka
f7104db26a KVM: Fix racy event propagation in timer
Minor issue that likely had no practical relevance: the kvm timer
function so far incremented the pending counter and then may reset it
again to 1 in case reinjection was disabled. This opened a small racy
window with the corresponding VCPU loop that may have happened to run
on another (real) CPU and already consumed the value.

Fix it by skipping the incrementation in case pending is already > 0.
This opens a different race windows, but may only rarely cause lost
events in case we do not care about them anyway (!reinject).

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:57 +03:00
Gleb Natapov
33e4c68656 KVM: Optimize searching for highest IRR
Most of the time IRR is empty, so instead of scanning the whole IRR on
each VM entry keep a variable that tells us if IRR is not empty. IRR
will have to be scanned twice on each IRQ delivery, but this is much
more rare than VM entry.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:57 +03:00
Gleb Natapov
6edf14d8d0 KVM: Replace pending exception by PF if it happens serially
Replace previous exception with a new one in a hope that instruction
re-execution will regenerate lost exception.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:56 +03:00
Marcelo Tosatti
54dee9933e KVM: VMX: conditionally disable 2M pages
Disable usage of 2M pages if VMX_EPT_2MB_PAGE_BIT (bit 16) is clear
in MSR_IA32_VMX_EPT_VPID_CAP and EPT is enabled.

[avi: s/largepages_disabled/largepages_enabled/ to avoid negative logic]

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:56 +03:00
Marcelo Tosatti
68f89400bc KVM: VMX: EPT misconfiguration handler
Handler for EPT misconfiguration which checks for valid state
in the shadow pagetables, printing the spte on each level.

The separate WARN_ONs are useful for kerneloops.org.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:56 +03:00
Marcelo Tosatti
94d8b056a2 KVM: MMU: add kvm_mmu_get_spte_hierarchy helper
Required by EPT misconfiguration handler.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:56 +03:00
Marcelo Tosatti
4d88954d62 KVM: MMU: make for_each_shadow_entry aware of largepages
This way there is no need to add explicit checks in every
for_each_shadow_entry user.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:55 +03:00
Marcelo Tosatti
e799794e02 KVM: VMX: more MSR_IA32_VMX_EPT_VPID_CAP capability bits
Required for EPT misconfiguration handler.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:55 +03:00
Andre Przywara
71db602322 KVM: Move performance counter MSR access interception to generic x86 path
The performance counter MSRs are different for AMD and Intel CPUs and they
are chosen mainly by the CPUID vendor string. This patch catches writes to
all addresses (regardless of VMX/SVM path) and handles them in the generic
MSR handler routine. Writing a 0 into the event select register is something
we perfectly emulate ;-), so don't print out a warning to dmesg in this
case.
This fixes booting a 64bit Windows guest with an AMD CPUID on an Intel host.

Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:54 +03:00
Marcelo Tosatti
2920d72857 KVM: MMU audit: largepage handling
Make the audit code aware of largepages.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:54 +03:00
Marcelo Tosatti
2aaf65e8c4 KVM: MMU audit: audit_mappings tweaks
- Fail early in case gfn_to_pfn returns is_error_pfn.
- For the pre pte write case, avoid spurious "gva is valid but spte is notrap"
  messages (the emulation code does the guest write first, so this particular
  case is OK).

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:54 +03:00
Marcelo Tosatti
48fc03174b KVM: MMU audit: nontrapping ptes in nonleaf level
It is valid to set non leaf sptes as notrap.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:54 +03:00
Marcelo Tosatti
e58b0f9e0e KVM: MMU audit: update audit_write_protection
- Unsync pages contain writable sptes in the rmap.
- rmaps do not exclusively contain writable sptes anymore.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:53 +03:00
Marcelo Tosatti
08a3732bf2 KVM: MMU audit: update count_writable_mappings / count_rmaps
Under testing, count_writable_mappings returns a value that is 2 integers
larger than what count_rmaps returns.

Suspicion is that either of the two functions is counting a duplicate (either
positively or negatively).

Modifying check_writable_mappings_rmap to check for rmap existance on
all present MMU pages fails to trigger an error, which should keep Avi
happy.

Also introduce mmu_spte_walk to invoke a callback on all present sptes visible
to the current vcpu, might be useful in the future.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:53 +03:00
Marcelo Tosatti
776e663336 KVM: MMU: introduce is_last_spte helper
Hiding some of the last largepage / level interaction (which is useful
for gbpages and for zero based levels).

Also merge the PT_PAGE_TABLE_LEVEL clearing loop in unlink_children.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:53 +03:00
Avi Kivity
3f5d18a965 KVM: Return to userspace on emulation failure
Instead of mindlessly retrying to execute the instruction, report the
failure to userspace.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:52 +03:00
Gleb Natapov
988a2cae6a KVM: Use macro to iterate over vcpus.
[christian: remove unused variables on s390]

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:52 +03:00
Gleb Natapov
73880c80aa KVM: Break dependency between vcpu index in vcpus array and vcpu_id.
Archs are free to use vcpu_id as they see fit. For x86 it is used as
vcpu's apic id. New ioctl is added to configure boot vcpu id that was
assumed to be 0 till now.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:52 +03:00
Gleb Natapov
1ed0ce000a KVM: Use pointer to vcpu instead of vcpu_id in timer code.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:52 +03:00
Gleb Natapov
c5af89b68a KVM: Introduce kvm_vcpu_is_bsp() function.
Use it instead of open code "vcpu_id zero is BSP" assumption.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:51 +03:00
Avi Kivity
d555c333aa KVM: MMU: s/shadow_pte/spte/
We use shadow_pte and spte inconsistently, switch to the shorter spelling.

Rename set_shadow_pte() to __set_spte() to avoid a conflict with the
existing set_spte(), and to indicate its lowlevelness.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:51 +03:00
Avi Kivity
43a3795a3a KVM: MMU: Adjust pte accessors to explicitly indicate guest or shadow pte
Since the guest and host ptes can have wildly different format, adjust
the pte accessor names to indicate on which type of pte they operate on.

No functional changes.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:51 +03:00
Avi Kivity
439e218a6f KVM: MMU: Fix is_dirty_pte()
is_dirty_pte() is used on guest ptes, not shadow ptes, so it needs to avoid
shadow_dirty_mask and use PT_DIRTY_MASK instead.

Misdetecting dirty pages could lead to unnecessarily setting the dirty bit
under EPT.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:50 +03:00
Avi Kivity
7ffd92c53c KVM: VMX: Move rmode structure to vmx-specific code
rmode is only used in vmx, so move it to vmx.c

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:50 +03:00
Avi Kivity
6a4a983973 KVM: Reorder ioctls in kvm.h
Somehow the VM ioctls got unsorted; resort.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:50 +03:00
Nitin A Kamble
3a624e29c7 KVM: VMX: Support Unrestricted Guest feature
"Unrestricted Guest" feature is added in the VMX specification.
Intel Westmere and onwards processors will support this feature.

    It allows kvm guests to run real mode and unpaged mode
code natively in the VMX mode when EPT is turned on. With the
unrestricted guest there is no need to emulate the guest real mode code
in the vm86 container or in the emulator. Also the guest big real mode
code works like native.

  The attached patch enhances KVM to use the unrestricted guest feature
if available on the processor. It also adds a new kernel/module
parameter to disable the unrestricted guest feature at the boot time.

Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:49 +03:00
Marcelo Tosatti
fa40a8214b KVM: switch irq injection/acking data structures to irq_lock
Protect irq injection/acking data structures with a separate irq_lock
mutex. This fixes the following deadlock:

CPU A                               CPU B
kvm_vm_ioctl_deassign_dev_irq()
  mutex_lock(&kvm->lock);            worker_thread()
  -> kvm_deassign_irq()                -> kvm_assigned_dev_interrupt_work_handler()
    -> deassign_host_irq()               mutex_lock(&kvm->lock);
      -> cancel_work_sync() [blocked]

[gleb: fix ia64 path]

Reported-by: Alex Williamson <alex.williamson@hp.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:49 +03:00
Marcelo Tosatti
60eead79ad KVM: introduce irq_lock, use it to protect ioapic
Introduce irq_lock, and use to protect ioapic data structures.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:49 +03:00
Marcelo Tosatti
64a2268dcf KVM: move coalesced_mmio locking to its own device
Move coalesced_mmio locking to its own device, instead of relying on
kvm->lock.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:49 +03:00
Marcelo Tosatti
9f4cc12765 KVM: Grab pic lock in kvm_pic_clear_isr_ack
isr_ack is protected by kvm_pic->lock.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:48 +03:00
Jan Kiszka
238adc7705 KVM: Cleanup LAPIC interface
None of the interface services the LAPIC emulation provides need to be
exported to modules, and kvm_lapic_get_base is even totally unused
today.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:48 +03:00
Liu Yu
06579dd9c1 KVM: ppc: e500: Add MMUCFG and PVR emulation
Latest kernel started to use these two registers.

Signed-off-by: Liu Yu <yu.liu@freescale.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:47 +03:00
Liu Yu
5b7c1a2c17 KVM: ppc: e500: Directly pass pvr to guest
Signed-off-by: Liu Yu <yu.liu@freescale.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:47 +03:00
Liu Yu
0cfb50e50d KVM: ppc: e500: Move to Book-3e MMU definitions
According to commit 70fe3af840.

Signed-off-by: Liu Yu <yu.liu@freescale.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:47 +03:00
Avi Kivity
105f8d40a7 KVM: Calculate available entries in coalesced mmio ring
Instead of checking whether we'll wrap around, calculate how many entries
are available, and check whether we have enough (just one) for the pending
mmio.

By itself, this doesn't change anything, but it paves the way for making
this function lockless.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:47 +03:00
Avi Kivity
596ae89565 KVM: VMX: Fix reporting of unhandled EPT violations
Instead of returning -ENOTSUPP, exit normally but indicate the hardware
exit reason.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:46 +03:00
Avi Kivity
6de4f3ada4 KVM: Cache pdptrs
Instead of reloading the pdptrs on every entry and exit (vmcs writes on vmx,
guest memory access on svm) extract them on demand.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:46 +03:00
Avi Kivity
8f5d549f02 KVM: VMX: Simplify pdptr and cr3 management
Instead of reading the PDPTRs from memory after every exit (which is slow
and wrong, as the PDPTRs are stored on the cpu), sync the PDPTRs from
memory to the VMCS before entry, and from the VMCS to memory after exit.
Do the same for cr3.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:46 +03:00
Avi Kivity
2d84e993a8 KVM: VMX: Avoid duplicate ept tlb flush when setting cr3
vmx_set_cr3() will call vmx_tlb_flush(), which will flush the ept context.
So there is no need to call ept_sync_context() explicitly.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:46 +03:00
Gregory Haskins
6b66ac1ae3 KVM: do not register i8254 PIO regions until we are initialized
We currently publish the i8254 resources to the pio_bus before the devices
are fully initialized.  Since we hold the pit_lock, its probably not
a real issue.  But lets clean this up anyway.

Reported-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Acked-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:45 +03:00
Gregory Haskins
d76685c4a0 KVM: cleanup io_device code
We modernize the io_device code so that we use container_of() instead of
dev->private, and move the vtable to a separate ops structure
(theoretically allows better caching for multiple instances of the same
ops structure)

Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Acked-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:45 +03:00
Gregory Haskins
787a660a4f KVM: Clean up coalesced_mmio destruction
We invoke kfree() on a data member instead of the structure.  This works today
because the kvm_io_device is the first element of the private structure, but
this could change in the future, so lets clean this up.

Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Acked-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:45 +03:00
Stephen Rothwell
2986b8c72c KVM: powerpc: fix some init/exit annotations
Fixes a couple of warnings like this one:

WARNING: arch/powerpc/kvm/kvm-440.o(.text+0x1e8c): Section mismatch in reference from the function kvmppc_44x_exit() to the function .exit.text:kvmppc_booke_exit()
The function kvmppc_44x_exit() references a function in an exit section.
Often the function kvmppc_booke_exit() has valid usage outside the exit section
and the fix is to remove the __exit annotation of kvmppc_booke_exit.

Also add some __init annotations on obvious routines.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:44 +03:00
Avi Kivity
6c8166a77c KVM: SVM: Fold kvm_svm.h info svm.c
kvm_svm.h is only included from svm.c, so fold it in.

Signed-off-by: Avi Kivity <avi@redhat.com>
2009-09-10 08:32:44 +03:00